To evaluate providers based on the health outcomes or the cost of care, one must attempt to evaluate dimensions of care which are strictly within the providers control. For instance, if a physicians treats two patients with breast cancer, but one patient has a more advanced form of breast cancer, one should take this difference into account. Patient comorbidities also affect the prognosis for a successful recovery from illness, as well.
One method to take into account the patient’s health conditions upon presentation at a provider’s facility is to use risk adjustment methods. Risk adjustment methods take into account factors such as patient demographics (e.g., age, gender), health status (e.g., prior diagnosis, current illness severity), prior utilizations (e.g., previous hospitalizations) and other factors to predict the expected outcome for a typical patient. Risk adjustment, however, is never perfect. A paper by Garber, MaCurdy and McClellen (1998) review some of the problems with using risk adjustment in the health care setting.
Lack of Precision
State-of-the-art risk adjustment methods based on diagnoses typically explain about 7% of the variation in medical expenditures from one year to the next. Although much of the remaining 93% may not be predictable, and thus is not a problem for insurance markets, a considerable fraction of it is likely to be predictable, in the sense that beneficiaries know it when they are choosing plans. For example, many intensive medical procedures such as elective joint replacements or cataract operations are largely predictable, in that an enrollee may be able to wait several months from the time that he or she suspects that the operation would be beneficial, so that the need for the procedure is known well in advance.
“…if intensive treatments and prior expenditures are included in risk adjustment methods, they introduce another incentive problem: An individual or a plan would be more willing to provide higher-cost treatments, because they would know that providing those treatments today would increase the payments they would receive in the future.”
“To the extent that expenditure differences in the future are known and Medicare compensates beneficiaries for these differences, then Medicare is redistributing wealth from low- to high-risk individuals rather than simply providing insurance. For example, if I know that I am likely to develop chronic lung disease, because it runs in my family or because I smoke, it might make sense for me to save money now in anticipation of the costs of hospitalization or higher costs of my insurance in the future. Such precautionary saving is expected when future events are certain and compensation through an insurance program is not available. Without risk adjustment, I might save more, because I would not expect as much compensation from Medicare over my lifetime; with risk adjustment, I would probably be better off, but those with lower risks of chronic illnesses—who would presumably receive fewer benefits or would contribute to the risk adjustment payments—might be worse off.”
- Alan M. Garber, Thomas E. MaCurdy, and Mark B. McClellan (1998) “Persistence of Medicare Expenditures among Elderly Beneficiaries,” Forum for Health Economics & Policy: Vol. 1: (Frontiers in Health Policy Research), Article 6.
What about coupling additional information, such as health risk assessment, EMR data, pharmacy claims, etc. with predictive modeling tools to risk adjust the patient population?
In order for risk to successfully shift to the provider, the providers need to have a sense that there is something they can do about the patient population. For docs that feel the risk adjustment isn’t accurate, why not focus on using similar metrics to determine gaps in care for known patients or populations? Eg. target all the diabetics that haven’t been in for an annual eye exam. The evaluation can be on the extent to which the risk shifted over time.